Adds ECL as a second supported implementation, specifically a statically
linked ECL. This is interesting because we can create statically linked
binaries, but has a few drawbacks which doesn't make it generally
useful:
* Loading things is very slow: The statically linked ECL only has byte
compilation available, so when we do load things or use the REPL it is
significantly worse than with e. g. SBCL.
* We can't load shared objects via the FFI since ECL's dffi is not
available when linked statically. This means that as it stands, we
can't build a statically linked //web/panettone for example.
Since ECL is quite slow anyways, I think these drawbacks are worth it
since the biggest reason for using ECL would be to get a statically
linked binary. If we change our minds, it shouldn't be too hard to
provide ecl-static and ecl-dynamic as separate implementations.
ECL is LGPL and some libraries it uses as part of its runtime are as
well. I've outlined in the ecl-static overlay why this should be of no
concern in the context of depot even though we are statically linking.
Currently everything is building except projects that are using cffi to
load shared libaries which have gotten an appropriate
`badImplementations` entry. To get the rest building the following
changes were made:
* Anywhere a dependency on UIOP is expressed as `bundled "uiop"` we now
use `bundled "asdf"` for all implementations except SBCL. From my
testing, SBCL seems to be the only implementation to support using
`(require 'uiop)` to only load the UIOP package. Where both a
dependency on ASDF and UIOP exists, we just delete the UIOP one.
`(require 'asdf)` always causes UIOP to be available.
* Where appropriate only conditionally compile SBCL-specific code and
if any build the corresponding files for ECL.
* //lisp/klatre: Use the standard condition parse-error for all
implementations except SBCL in try-parse-integer.
* //3p/lisp/ironclad: disable SBCL assembly optimization hack for all
other platforms as it may interfere with compilation.
* //3p/lisp/trivial-mimes: prevent call to asdf function by substituting
it out of the source since it always errors out in ECL and we hardcode
the correct path elsewhere anyways.
As it stands ECL still suffers from a very weird problem which happens
when compiling postmodern and moptilities:
https://gitlab.com/embeddable-common-lisp/ecl/-/issues/651
Change-Id: I0285924f92ac154126b4c42145073c3fb33702ed
Reviewed-on: https://cl.tvl.fyi/c/depot/+/3297
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Reviewed-by: eta <tvl@eta.st>
CCL refuses to compile anything with undeclared ignored variables, so we
need to be more verbose here.
Change-Id: I9bf32e0bc303716d3cd6fe7b525d1434062d69eb
Reviewed-on: https://cl.tvl.fyi/c/depot/+/3348
Tested-by: BuildkiteCI
Reviewed-by: grfn <grfn@gws.fyi>
Found this typo because CCL is really particular about everything
related to declare.
Change-Id: I6d4615c1df7c9d45722e85fa82ebdd094273205d
Reviewed-on: https://cl.tvl.fyi/c/depot/+/3347
Tested-by: BuildkiteCI
Reviewed-by: grfn <grfn@gws.fyi>
Turns out this is an annoying thingy you sometimes to implement
independently from formatting an entire timestamp, so we expose it for
reuse.
Change-Id: I11de2823eb03849ea78fc79e2f546e413882930f
Reviewed-on: https://cl.tvl.fyi/c/depot/+/3163
Tested-by: BuildkiteCI
Reviewed-by: grfn <grfn@gws.fyi>
Reviewed-by: tazjin <mail@tazj.in>
Instead of bothering with local-time's feature rich timezone support we
just pass the offset as an integer and render it ourselves.
Change-Id: I1df2d02153e3ef21ae3b2871ad6ef57d0f6eff86
Reviewed-on: https://cl.tvl.fyi/c/depot/+/2423
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Reviewed-by: glittershark <grfn@gws.fyi>
Display the history of an issue (which currently is just opening and
closing) inline with the issue's comments on the issue show page
Change-Id: Id167bceef765cb4c24e86983d1dcd6624d0e5956
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1497
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Read the port and data directory from environment variables, in
preparation for deploying as a systemd unit to Whitby
Change-Id: I066dced7b7926b6bdc77132d13a4da6c886b20e8
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1338
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Initial commit for Panettone, a very simple issue tracker for TVL. In
its current state this launches a web server with authenticates with our
ldap server, and supports listing and creating issues via static html
pages and simple forms.
We've been needing an issue tracker for a while now, but none of the
options out there seem very good - or there are some good ones, but
they're AGPL licensed and we don't want to deal with them. Rather than
muck around with Trac or Bugzilla, we've decided to write our own.
Change-Id: I704f0996d15199329bbd5450f3d959046bf13973
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1337
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Add a function to klatre format a timestamp using dottime
Change-Id: I24d8d91f49f352b606f44834f7229ab55b55afa0
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1344
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
Reviewed-by: kanepyork <rikingcoding@gmail.com>
Add Klatre, a grab-bag common lisp utility package, including
definitions for `comment`, `posp`, `chunk-list`, and `mapconcat`.
The name traces its lineage back to Abseil, a similar grab-bag utility
library for C++ - abseiling is what you do to go down a route after
you're done climbing it, and klatre is norwegian for "to climb"
Change-Id: I5efd91d8af827883679ce1a2eed3229b28e082ac
Reviewed-on: https://cl.tvl.fyi/c/depot/+/1346
Tested-by: BuildkiteCI
Reviewed-by: tazjin <mail@tazj.in>
This enables arbitrary DNS lookups (with the caveat that RRDATAs are
currently not deserialised into a record-type-specific format).
An error condition has been defined for error-responses from the HTTP
server which provides interactive restarts for attempting a new call
with different parameters.
Implements support for the compresion scheme used in binary DNS
messages.
This makes it possible to decode messages entirely, but not yet
actually resolve the labels to their "real" values.
All qnames are stored with file-offsets pointing at the position at
which their reading started, which enables the implementation of a
function to resolve pointers internally.
Adds a struct that represents QNAMEs, tracks the stream offset at
which the QNAME parsing began and makes it possible to resolve
pointers inside of the QNAME.
Note that resolving pointers needs to happen *after* the call to
lisp-binary currently. It might be possible to implement this inside
of lisp-binary in the future by switching on the top two bits of the
qname field, but since this is happening *inside* of a reader function
I'm not currently sure how to implement it.
This uses lisp-binary to define serialisation types for the DNS
messages defined by RFC 1035.
Currently the compression scheme used for QNAMEs is not supported,
hence deserialisation of even simple records fails after the header
and question sections are read.
Adds a package definition file and moves the current client into
client.lisp
Note that the client is not working at all at this commit as this is a
work-in-progress snapshot.
This includes very barebones support for querying TXT and MX records
right now. The returned structure is not turned into a more convenient
format and error handling is, well, NIL.
